A Framework for Planning a Unified Wired and Wireless ICT Infrastructure

A Framework for Planning a Unified Wired and Wireless ICT Infrastructure
of 6
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
    Aalborg UniversitetA Framework for Planning a Unified Wired and Wireless ICT Infrastructure Riaz, Tahir; Nielsen, Rasmus Hjorth; Pedersen, Jens Myrup; Prasad, Neeli R.; Madsen, OleBrun Publication date:  2007 Document Version  Publisher final version (usually the publisher pdf)Link to publication from Aalborg University Citation for published version (APA):  Riaz, M. T., Nielsen, R. H., Pedersen, J. M., Prasad, N. R., & Madsen, O. B. (2007). A Framework for Planning aUnified Wired and Wireless ICT Infrastructure. Paper presented at The 10th International Symposium onWireless Personal Multimedia Communication (WPMC 07), Jaipur, India. General rights Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright ownersand it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. ? Users may download and print one copy of any publication from the public portal for the purpose of private study or research. ? You may not further distribute the material or use it for any profit-making activity or commercial gain ? You may freely distribute the URL identifying the publication in the public portal ? Take down policy If you believe that this document breaches copyright please contact us at providing details, and we will remove access tothe work immediately and investigate your claim.Downloaded from on: august 08, 2015  A Framework for Planning a Unified Wired and WirelessICT Infrastructure Tahir M. Riaz, Rasmus H. Nielsen, Jens M. Pedersen, Neeli R. Prasad, Ole B. Madsen Networking and Security Section, Center for Network Planning (CNP)Center for TeleInFrastruktur (CTiF), Department of Electronic SystemsAalborg University, Niels Jernes Vej 14, 9220 Aalborg, Denmark e-mail:  { tahir, rhn, jens, np, obm }  Abstract —The increase in the use of information and communicationtechnology (ICT) has pushed the existing access networks to theirlimits. Whole new access networks are currently being deployed andare expected to take full advantage of the already started synergyof services converging on to one network. Through a brief survey of synergy and technology trends, it is concluded that a future networkwill complementary use of wired and wireless technologies. In thiscontext the paper proposes a framework for planning of unified wiredand wireless ICT infrastructures. The framework includes differentinput parameters of relevance for the planning and implementation,which also include a step-wise implementation plan. Planning methodsfor wired and wireless planning is presented and a simplified large-scale case study is conducted to verify and illustrate the use of theframework.  Keywords —Wireless Access, WiMAX, FTTH, Broadband Networks,Network Planning, ICT infrastructure, Modeling I. I NTRODUCTION Most of the world have experienced a large increase in the useof information and communication technology (ICT) for almostevery possible purpose (for a Danish perspective - see [1 fromelectronic mail, online shopping, and gaming to citizen service,remote control, and telemedicine. This development has triggeredthe penetration of increasingly faster broadband connections [2],and thereby greatly increased the load on the existing ICT network infrastructure and pushed it towards its limits.Upgrades are currently being carried out on the access network.unlimited bandwidth. And so, for the first time in 100 years,a completely new media is about to be deployed in the wiredaccess networks in a large number of countries on the initiativeof governments, telcos and other market players. At the sametime new wireless networks are also being deployed for both fixedand mobile access as the development in wireless technologies isemerging into still higher bandwidths and ranges.Most of the old telcos are offering both wired and wirelessnetwork coverage. Currently the wired coverage is mostly offeredthrough the old copper-based PSTN or CATV networks, wheremost, if not all, of the PSTN networks, used for POTS and ISDN,are owned by the former telephony monopolies, which have alsotaken a large share of the CATV networks, while the remainingCATV networks are owned by local or regional communities.The CATV networks are closed for competition, while the POTSnetworks are regulated and thus open for competitors, who canlease the raw copper and co-locate equipment on existing centraloffices or offer a connection through bit stream access, wherethe user traffic is delivered to a point in the operator’s network.The network connection delivered by xDSL and even though highbandwidths can be delivered it is limited to the customers close tothe central office.New deployments by market-newcomers are being carried outboth with respect to wireless and wired networks. Most noticeableis the ongoing effort by newcomers to deploy Fiber to the Home(FTTH), but WiMAX implementations in larger cities are alsomore and more often announced. A case study of broadbandinfrastructure in the region of Western Greece can be found in [3]where both wired and wireless solutions are discussed separately.For a majority of the new FTTH network operators the end goalis clear: To provide a 100% FTTH coverage. The time horizon forthis goal is often relatively short 1 .It must be expected to be of the utmost importance for FTTHnewcomers to have a rather high penetration rate in order to keepthe customers satisfied and to win customers from day one. Dueto the fundamental nature of digging down fiber, where especiallythe digging is a time consuming task, it can be relevant to considerothers steps and approaches. Such could include buying in onexisting wired or wireless networks, leasing or roaming on existingnetworks, or to consider other technologies than fiber, e.g. wireless,for a transition phase and/or as a complementary technology forthe end scenario.From the possibilities listed, the latter is the most appealing. Notonly will it be much faster to deploy a wireless network, but it willalso balance out the advantage of being able to offer a full packagecontaining wired and wireless coverage, currently possessed byexisting network operators. However, taking this approach willrequire an even more significant planning task than the planningof a wired network. Not only will the end goal be extremelyimportant, but also the individual steps of going there will be of high importance.In relation to the just mentioned considerations, this paper willpresent a framework for the stepwise planning of a unified wiredand wireless network for FTTH newcomers. The framework willbe useful as a more structured approach to network planning andwill make it possible to further enhance the inclusion of moresophisticated planning methods along with different aspects, suchas an overall business plan.The remaining paper will first include a survey of the currenttrends within technologies and network synergies to further setupa basis for the subject considered. In section III the inputs fora planning model are considered, followed by description of theindividual implementation steps in section IV. Afterwards theplanning model will be surveyed and a case study conducted,followed by results, conclusion and discussion.II. S YNERGY AND  T ECHNOLOGY  T RENDS Traditionally, the different ICT infrastructures have been dis-persed into several big and small segments, ranging from fixedtelephony, cellular networks, broadcast wireless networks: TV andRadio, cable TV, surveillance etc. Over time these networks havebeen diverging, converging, but mostly coexisting more or lessindependently of each other. In general the networks have beenservice-oriented - one network for one service and the networkshave been specified in order to support for the services traversingit. The still increasing use of the Internet and thus IP (Internet 1 A Danish FTTH newcomer [4] is expecting to deploy 30,000 km of ducts filled with 18 million km of fiber within 6 years (around 14 kmduct/day and 8,220 km fiber/day). 375  Protocol) has introduced and opened for the basic possibility of synergy, where all services can use the same network. Now thequestion is; if a network, using a single protocol is available andis able to offer all kind of services then why are all these differentkinds of networks necessary, why not just have one network thatis able to fulfill all the requirements of the services. The answer isthat it all mainly come down to the limitation of technology itself.When considering the future, the technologies in front withinwired and wireless respectively are; FTTH and 4G wireless net-work. FTTH will be replacing the cooper networks, and 4G willevolve from existing wireless technologies. FTTH promises ascalable incomparable bandwidth, while 4G basically will be anetwork of networks, capable of using multiple wireless tech-nologies simultaneously [5], [6]. There are two main wirelessbroadband technologies generally known as: 3G and WiMAX. 3Gis a successor of the previous cellular networks e.g. GSM, iDEN,IS-95, GPRS etc., and WIMAX (802.16x) is the successor of WiFi(802.11x). The latest standards for WiMAX are 802.16d-2004 forfixed access with a typical coverage of up to 50 km, and 802.16e-2005 which includes mobility and provides 70 Mbps within a rangeof 3 to 8 km [7].Whichever technology will be dominating in the wireless seg-ment it is clear that it will be an integrated part of a wiredinfrastructure. Like today most of the users already have wiredand wireless networks at the same time for different services, andin the future this phenomenon will remain alive unless some veryunusual breakthrough hits which totally outperforms the wiredsegment. The infrastructure of the future is thus not be eitherwired or wireless, but rather wired and wireless as complementarytechnologies for the greater good of the user.III. M ODEL  I NPUTS Before considering the actual implementation steps, it is neces-sary to consider the physical constraints forming the inputs for themodel.This include: Technology, geography, implementation and cus-tomers. In the following the inputs will described along with theindividual parameters of importance for the inputs.  A. Technology An essential parameter is the bandwidth offered per customer,as it explicitly influence the lifetime of a given technology. Thebandwidth should be sufficient for a time frame of at least thedepreciation period of the equipment. Influence on bandwidthfrom distance, interference, overhead, shared media etc., especiallyrelevant for wireless technologies, should be parameterized and theeffects studied for the considered area to give an exact measure of the impact on a given technology.The coverage of the access technology is also an importantparameter as it directly influence the number of central offices orbase stations in an area. The lower the coverage range the higherthe number of such points to reach all customers.For a wireless network, the mobility must also be consideredalong with the complete cost for deploying the technology includ-ing both the initial and the operating expenses included.A lot more parameters exist and all relevant parameters must beconsidered. When surveying technologies, it is not likely that onetechnology will be the better option with respect to all parameters,as parameters will contradict with each other e.g. bandwidth versusmobility. The parameters must thus be weighted according to theirrelevance for the specific area, to choose the best technology. 1) Network Architecture:  With respect to technology it is alsorelevant to consider the network architecture used. In this paper, ahierarchical network architecture is proposed. The architecture de-fines how the different network elements should be interconnected.The hierarchical level can be found as organized in various waysand depends on the technology and the network requirements. NTNT NT NT NT NT MN MN DNDN AN ANDistribution Network Access Network  Backbone Network  Fig. 1. The proposed network architecture. The proposed architecture, shown in Figure 1, ensures redun-dancy at all the hierarchical levels. The distribution nodes (DN) areconnected using two independent paths to two different main nodes(MN). The network must be able to survive not only in case of linefailures but also in case of node failures. In the next level of thehierarchy the access nodes (AN) (or central offices as used in thispaper) are connected to the distribution nodes. The access nodesalso include base stations for the wireless access. The principle isthe same as the previous hierarchical levels; all the access nodesare to be connected to the upper hierarchical levels with line andnode independent paths.  B. Geography The geographical area to be planned is of course a very importantparameter and together with the customers, this is making up thedemography. The geography is also of importance when choosingtechnology and considering the price. For the wired part mostlythe price is influenced by the geography as the digging cost can begreatly increased if the underground is made up of rock or if a lot of rivers and streams are to be crossed. For a large geographical areawith a low number of scattered customers, the maximum range canalso be subject to consideration. For the wireless part the geographycan influence parameters such as attenuation, fading and reflectioneffects etc., which is most likely to be of importance if consideringdense urban areas, areas with mountains or forests etc. As with thewired technology, cost will be influenced by the geography, mainlybecause of an increased number of base stations. C. Implementation The inputs from the implementation, mainly cover boundariesset up by the business case. These inputs include the time framefor building the new infrastructure, the resources available and thespread of the implementation.  D. Customers Providing services to customers is the basic condition for theexistence of a network. The type of customer, and the differencebetween customers of the same type, vary greatly; e.g. from sensorsto HDTV consumers.Each customer is prioritized in order to decide on the orderof connectivity - the higher the priority, the sooner the customeris connected. The priority is calculated as a function of variousparameters, illustrated through Figure 2. 376  CustomerParameter NWeight PriorityRedundancyProfit of SubscriptionCost of ReachBandwidthContractFig. 2. Setting up parameters for each customer and calculating the priority. Many parameters exist, but they are not equally important. If thecustomer has or is part of a contract specifying an actually latesttime of connectivity, this should obvious increase the priority of the customers.Providing redundancy comes at an extra cost, which does notonly cover the extra line or antenna needed to give the customertwo connections. Redundancy has to be insured all way throughthe upper level networks by having nodes and lines independencebetween the two paths on the logical as well as the physical layer.The impact on priority is thus a matter of time. If a fully redundantbackbone has been implemented, the redundancy will be balancedwith the extra subscription fee charge for the service. If, however,at the given time redundancy has not been implemented, the costis greatly increased as not only the extra access line has to beestablished, but also lines and nodes in the upper level networks.IV. I MPLEMENTATION  S TEPS The implementation of a new ICT infrastructure will be dividedin three steps: Operational, Tactical, and Strategic. In each step,the synergy between the different network levels is utilized in full(see e.g. [8] for the significance of synergy in the access network).Whenever a certain trace for one of the network levels has beendug up, the capacity needed in this trace for the other network levels, is included as well (at least in term of empty tubes).  A. Operational Step The operational step includes the essential task of building acore infrastructure to be the backbone of the network about to bebuilt.Full redundancy is not needed at this stage, why focus is onconnectivity. However, the connectivity is provided in order toeventually support a structure (such as a grid [9]) that can providegood properties with respect to Structural Quality of Service(SQoS) [10]. The number of fibers for each line are deployed in amagnitude able to support the traffic in a time frame at least equal tothe span of the three steps, when considering the rapid developmentin capacity available through upgrade of end-equipment.High priority customers will also be connected at this stepand depending on the customer priority the connectivity shouldbe offered as a wired or a wireless connection. If the highpriority customers are requiring wired redundancy, the numberof central offices are kept down by establishing the longer wiredconnection for a part of the customers. A number of base stationsare established to increase the number of customers. The basestations are located to maximize the number of customers reached,considering the bandwidth available. For the customers requiringredundancy, the secondary connection is offered through wirelessand this is also included in the location of the base stations.  B. Tactical Step The core infrastructure is upgraded for redundant connections,to end up in the structure decided on. If the core lines werenot initially setup to support the traffic load in this step, end-equipment is upgraded to support the current and future load. Acomplete wireless network is deployed in order to reach the wholearea. Customers are connected to the wireless network, as long assufficient bandwidth is available. Priority customers are offeredredundancy through either the wired or the wireless network.Customers from base stations, which are getting overloaded, shouldbe given higher priority in order to offload the wireless network. C. Strategic Step In the strategic step all customers are gradually connected witha wire with respect to their priority and bandwidth considerations,for their primary connection, can be discarded.Wired redundancy is established to all customers requiring it.The wireless network is now present as a redundant connection.All customers will thus have two independent connections andsome customers will have three independent connection, virtuallyeliminating all downtime. Furthermore the wireless connection willstill function as a dedicated access network for mobile users, andwill thus offer extra service to the customers.V. P LANNING  M ODEL To plan a network a planning model is essential. There are twomain categories, namely wireless planning and wired planning. Inthis paper the focus is on an overall planning rather than a detailedplanning, why the planning will not focus on strict optimizationsince the aim is to give an overall overview of a combined wiredand wireless network infrastructure. Traditionally, planning hasbeen carried out manually in an ad-hoc manner and wired andwireless infrastructures have been planned separately.  A. Wired Planning For the wired part a GIS based automated planning model, asproposed in [11], will be employed, which gives some systematicapproaches to network planning.Planning the wired part is relatively difficult when planning atlarge scale. In theory wires can be placed anywhere, but in reality itis not the case. Buildings, houses, mountains etc. can be obstacleswhen deploying the cables. One of the planning requirements is tofind the potential places where central offices can be located andwhere cables can be placed confronting minimum obstacles. Whenlooking at the real world the most used sites are road networks.Roads are means of interconnecting places for transportation andoften have a shortest distances. For the modelling purpose roadscould be a better choice for the potential sites for cable placement.Today, GIS data is easily available for many parts of the world.GIS data provides a road network in digital form.  B. Wireless Planning For the wireless part GIS information is also used to determinethe number of base stations needed and two main parametersare considered: coverage and bandwidth.An overall estimation isrequired to find out how many base stations are needed. Thelocations for the base stations and coverage are determined throughan algorithm containing the following steps.Step 1For the starting step, a square/cell is created coveringthe considered area. This can be done using two extremepoints of a square; either bottom right and top left orvice versa. These two reference points are also useful todetermine the number of customers and the locations forthe base stations.Step 2The squares are split into two smaller cells, and thissplitting function continues recursively until the distanceparameter is satisfied. In addition, the split predecessorcell is removed. 377  Step 3The cells are further split if needed along with satisfyingthe number of customers in the given cells. Since thebase station has a limited bandwidth capacity available,the size of the cell is mainly dictated by the number of customers within the cell.Step 4The cells with a low number of customers which arewithin the coverage of neighbor cells are removed - theneighbor cells should have sufficient capacity to includethe customers from the cell which is being removed.The model does not consider other parameters like attenuation,fading and reflection effect etc.VI. C ASE  S TUDY To investigate the usefulness of the model a case study has beenconducted and will be described in the following. The purpose of the case study is also to further illustrate the use of the model.  A. Model Inputs For the case study, the inputs for the strategic planning model areneeded. The following determines the inputs based on the generalframework given in section III. 1) Technology:  (Wired)  - FTTH access technology is used forthe wired network and it is considered to be offered as a homerunsolution giving 1 dedicated fiber from the customer to the centraloffice.Bandwidth is considered to be virtually unlimited and canalready be offered as 100 Mb/s or 1 Gb/s. The homerun is alsovery scalable, as multiple wavelengths can be used later on toincrease the bandwidth capacity of a single fiber. Furthermore thereach can easily be up to 10 km and further increased by choosingquality components and carefully handling the fiber and splices orin extreme cases to add in optical amplifiers. (Wireless)  - For the wireless part, a WiMAX 802.16e solutionhas been chosen. 802.16e is the latest standard and is already beingdeployed in many parts of world.WiMAX 802.16e offers a bandwidth in the order of 70 Mb/sper base station and the reach is up to 8 km, which is decrease to5 km considering the bandwidths needed.For the case study a required bandwidth of 10 Mb/s per customerhas been considered together with a 10% penetration and a 10%utilization, adding up to 700 potential customers per base station. 2) Geography:  As a case study area, the county of North Jutlandhas been chosen. The county [12], with a population of 495,068,is one of the most rural areas in Denmark covering 6,173 km 2 ,giving a population density of only 80 citizen per km 2 .This region does not setup special geographical considerations tobe taken. However, if considering the higher level networks and lineindependence, the fjord traversing the region can put up challenges.As Denmark is a rather flat terrain country, meaning it does nothave higher elevation such as mountains etc., wireless planning isdone without any consideration to other parameters than reach andbandwidth. 3) Implementation:  The time frame of the case study performedis expected to be 15 years, with the operational step to becompleted in 3 years, the tactical step in 5 years and the strategicstep in 15 years. This is considered together with a full spreadimplementation, meaning a full roll-out of all central offices andbase stations. It is presumed that the economics and workforce willbe available. 4) Customers:  For the case study, a full range of customer typesis not considered. The customer types considered are types wheredirect interaction is taking place with human beings. This choicehas not been made because the authors neglect the importance of control of power plants or traffic light, but first of all because suchtasks are likely to setup demands more suitable for a dedicatednetwork and secondly because the attended audience of this paper(FTTH newcomers) are not focusing on such types of customers.Based on the former considerations, the different customertypes can be defined and prioritized according to the parametersdescribed in section III-D. (Public)  - public institutions are considered high priority cos-tumers, as they are likely to have contractual agreements of delivering of network coverage. They can be expected to be onlong term leases and to have a well-defined usage pattern. Publicinstitutions are considered redundancy-dependent, as e.g. hospitalsare included in this category. For the case study area this amountsto 2,147 potential customers. (Business)  - businesses are also considered high priority cus-tomers. For certain large companies, contractual obligations canexist e.g. to connect branches of a company. Redundancy must alsobe considered for this category especially when considering theincreasingly use of IT in businesses, not to mention e-commerce.The business category includes 5,599 customers. (Private)  - private customers are by large the biggest amountof the customers, however, they are only considered at a mediumpriority. First of all the penetration rate is not easily determinedas no contracts are likely to be agreed upon before start of thedeployment. Related to this, the private customers are not likelyto agree upon long term leases with the operator and they aremore likely to change to another operator if a good offer isavailable. Lastly, the individual bandwidth requirement for this typeof customers is low compared to a business or public institutions.This category includes the remaining 284,016 customers in thearea.  B. Implementation Steps For the case study the implementation steps for the deploymentof a combined wired and wireless infrastructure will be as listedin the following. 1) Operational Step:  In this step every public institutions andbusiness will be offered connection through FTTH. This is enoughto fulfill the requirements with respect to bandwidth.Furthermore, a full wireless coverage is offered through thedeployment of a full range of WiMAX base stations. The basestations are fed by a single fiber wired connection. All customersare offered wireless connectivity and the WiMAX network will alsobe working as a second redundant connection for public institutionsand businesses. 2) Tactical Step:  The tactical step connects all customers withina radius of 2 km from central offices with a FTTH connection. Atthe same time all public institutions are given a wired redundantconnection. The same is the case for the WiMAX base stations. 3) Strategic Step:  In the final step all customers are offeredconnectivity through a FTTH connection. The business customersare given redundancy through a wired connection. The WiMAXnetwork will then no longer have any primary customers and willfunction as a network used for mobile users in the area. As todayit is becoming more common to use load balancing routers wheremultiple access connection can be used for backup and sharing.As such the two networks can both offload each other and be eachother redundant path, seamless to the customer. C. Results A complete unified wired and wireless network was planned forthe case study area. The number of customers along with the fiberand digging needed can be seen from Table 1 for the individualimplementation steps. The results for estimation the number of basestations need are illustrated in Figure 3. 378
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks